Conference paper Open Access
Potapczyk, Tomasz; Przybysz, Pawel; Chochowski, Marcin; Szumaczuk, Artur
{ "publisher": "Zenodo", "DOI": "10.5281/zenodo.3525498", "language": "eng", "title": "Samsung's System for the IWSLT 2019 End-to-End Speech Translation Task", "issued": { "date-parts": [ [ 2019, 11, 2 ] ] }, "abstract": "<p>This paper describes the submission to IWSLT 2019 End- to-End speech translation task by Samsung R&D Institute, Poland. We decided to focus on end-to-end English to German TED lectures translation and did not provide any submission for other speech tasks. We used a slightly altered Transformer architecture with standard convolutional layer preparing the audio input to Transformer en- coder. Additionally, we propose an audio segmentation al- gorithm maximizing BLEU score on tst2015 test set.</p>", "author": [ { "family": "Potapczyk, Tomasz" }, { "family": "Przybysz, Pawel" }, { "family": "Chochowski, Marcin" }, { "family": "Szumaczuk, Artur" } ], "type": "paper-conference", "id": "3525498" }
All versions | This version | |
---|---|---|
Views | 404 | 404 |
Downloads | 309 | 309 |
Data volume | 32.4 MB | 32.4 MB |
Unique views | 357 | 357 |
Unique downloads | 282 | 282 |